频域宽带下视多波束三维声成像算法

Broadband downward-looking multi-beam 3D imaging algorithm in frequency domain

为解决下视多波束三维成像声呐跨航向分辨率差、旁瓣级高的问题,提出一种频域宽带下视多波束三维声成像算法。该算法将阵元域回波信号通过二维傅里叶变换直接变换到波数-频率域,之后进行深度向匹配滤波,同时选取带宽内的频点进行频域宽带信号的相位补偿,再进行二维逆傅里叶变换回阵元域,实现跨航向-深度向成像处理。沿航向采用时域合成孔径方法,得到三维图像结果.计算机仿真和海上实验结果表明,相比于常规的下视多波束三维成像声呐处理算法,改进的方法跨航向主瓣变窄,角分辨率明显提高,旁瓣降低10~30 dB。相比于传统基于分块DFT的频域宽带波束形成,改进的频域宽带方法避免了跨航向处理中的时域分块插值和多次变换域,复数加法和乘法次数显著降低。改进的频域宽带下视多波束三维声成像算法可以有效地获取水下三维高分辨图像,抑制虚影,明显改善成像质量。同时,计算效率高,更适合于实时成像系统实现。

A broadband downward-looking multi-beam 3 D imaging algorithm in frequency domain is proposed.Echo in array-element domain is transferred into wave number-frequency domain through 2 D Fourier Transformation.Vertical match filtering and phase compensation of broadband signal are subsequently applied.The result is then transferred into array-element domain through 2 D Inverse Fourier Transformation to complete the across-track-vertical imaging processing.Synthetic aperture method is used for along-track processing to obtain the 3 D image.Computer simulation and sea trial results demonstrate that across-track resolution has obvious improvement and the side lobe level is significantly reduced of 10~30 dB compared with the conventional narrowband downward-looking multi-beam 3 D imaging algorithm.Compared to the traditional broadband beamforming,the computational complexity of the presented method is significantly reduced,which is more suitable for real time application.